Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
5722406 | Journal of Affective Disorders | 2017 | 9 Pages |
â¢There is interest in GPs providing suicide prevention to at-risk patients.â¢We conducted a systematic review and meta-analysis of these studies.â¢We found no evidence of reductions in suicide ideation, self harm or suicide attempt.â¢There was some evidence of reductions in death by suicide.
BackgroundThe aim of this review was to assess whether suicide prevention provided in the primary health care setting and delivered by GPs results in fewer suicide deaths, episodes of self-harm, attempts and lower frequency of thoughts about suicide.MethodsWe conducted a systematic review and meta-analysis using PRIMSA guidelines. Eligible studies: 1) evaluated an intervention provided by GPs; 2) assessed suicide, self-harm, attempted suicide or suicide ideation as outcomes, and; 3) used a quasi-experimental observational or trial design. Study specific effect sizes were combined using the random effects meta-analysis, with effects transformed into relative risk (RR).ResultsWe extracted data from 14 studies for quantitative meta-analysis. The RR for suicide death in quasi-experimental observational studies comparing an intervention region against another region acting as a “control” was 1.26 (95% CI 0.58, 2.74). When suicide in the intervention region was compared before and after the GP program, the RR was 0.78 (95% CI 0.62, 0.97). There was no evidence of a treatment effect for GP training on rates of suicide death in one cRCT (RR 1.07, 95% CI 0.79, 1.45). There was no evidence of effect for the most other outcomes studied.LimitationsAll of the studies included in this review are likely to have a high level of bias. It is also possible that we excluded or missed relevant studies in our review processConclusionsInterventions have produced equivocal results, which varied by study design and outcome. Given these results, we cannot recommend the roll out of GP suicide prevention initiatives.